ketching longer-term plans.
--
Pierre Ynard
on/vnd.adobe.flash.movie :/
--
Pierre Ynard
ie version of them; so that's
ridiculous to me.
Thanks for your consideration on this topic,
--
Pierre Ynard
-- System Information:
Debian Release: bullseye/sid
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (cha
ration on this issue,
--
Pierre Ynard
-- System Information:
Debian Release: bullseye/sid
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init:
Package: locales
Version: 2.31-9
Severity: normal
Dear Maintainer,
I use the fr_FR.UTF-8 locale. After upgrading from glibc 2.30-8 to
2.31-9, the output of the `date` command without arguments significantly
changed.
So far and with 2.30-8 still:
% date
lundi 8 février 2021, 20:31:43 (UTC+0100)
Package: libdvd-pkg
Version: 1.4.2-1-1
Severity: normal
Dear Maintainer,
My use case is the simple one, all I want is to play some DVDs with my
media player of choice. All I need is the libdvdcss2 binary package.
I don't need libdvd-pkg and all the build tools that come with it for
that, so
Package: libdvd-pkg
Version: 1.4.2-1-1
Severity: minor
Dear Maintainer,
/usr/share/doc/libdvdcss2/README contains this section:
> Building and Installing libdvdcss
> =
>
> See the INSTALL file.
However there is no /usr/share/doc/libdvdcss2/INSTALL or any
> And, I can't see standalone tune2fs even in unstable:
>
> $ apt-file find /bin/tune2fs
> android-sdk: /usr/lib/android-sdk/tools/bin/tune2fs
/sbin/tune2fs is shipped by e2fsprogs.
--
Pierre Ynard
sprotect/system in
the list is the right approach, that would be a very specific fix.
--
Pierre Ynard
please help and:
- look into this issue about FUSE filesystem processes getting killed
at shutdown?
- advise about anything that I might have failed to understand about
the unmounting issue?
- advise about FUSE network filesystem types and how to detect them?
Thanks,
--
Pierre Ynard
-a or -y in the
general case; hopefully they are at least silently ignored by most
implementations, aren't they?
--
Pierre Ynard
ot; to "right before" and add something like: "You
should only need this to handle issues with Wake-on-LAN or network
filesystems."
--
Pierre Ynard
> Well, let's start with something. What about this patch, that reuses
> /etc/default/halt? I believe it is, maybe not most principal, but
> solution to bug at hand.
Looks good to me
--
Pierre Ynard
sck, which
would then be checked by a script in runlevel 0 or 6 that calls tune2fs
on the relevant filesystems.
We could use that script any time we want (like in runlevel S too) to
print deprecation warnings to assist in the transition.
--
Pierre Ynard
ed that the machine will not hang for waiting on a
> process that will not terminated.
Considering that you should be able to do this in the way I mentioned
above, I don't think it's necessary to add such a mechanism - but maybe
I misunderstood your problem.
--
Pierre Ynard
bs | grep -q network.target && exit 0
> fi
Isn't that check adequate already? How should it be modified if not?
Thanks,
--
Pierre Ynard
if we
don't want to expand initscripts knowledge about every problematic
network filesystem, considering that other packages can't edit
/etc/default/halt, it might be even better to let them drop files
somewhere, maybe in /run/network, to signal this.
--
Pierre Ynard
c/default/halt, as requested in #703844.
Actually, #696910 suggests that NETDOWN could also be disabled if iSCSI
is detected.
--
Pierre Ynard
on whether they still experience the issue?
Dear NM maintainers, could you please look at this and tell us if you
think this issue still persists today? And help us figure out how
dependencies or sendsigs exception should be adjusted to resolve this?
Thanks,
--
Pierre Ynard
ted by
> protocol
Thanks,
--
Pierre Ynard
, so I'm closing this.
Thanks,
--
Pierre Ynard
by bootclean.sh
because it's a directory. I'm going to assume that's what happened then.
Nowadays /run is a tmpfs so that issue is mostly moot.
I suppose we could still improve that locking or make sure stray locks
are properly removed before mountnfs operations are started.
--
Pierre Ynard
should honor NETDOWN, and source
it from /etc/default/halt. And then NETDOWN should be documented in
/etc/default/halt, as requested in #703844.
--
Pierre Ynard
roblem.
--
Pierre Ynard
an alternative interface to creating a /forcefsck file would have to
be used, one that doesn't reside on the filesystem itself and can be set
and cleared for individual filesystems separately - for example, the
filesystem internal data.
--
Pierre Ynard
d by this issue and if they would
allow to address it. Ted, what do you think about this?
--
Pierre Ynard
uot; flag should
definitely call /sbin/shutdown -h -H and always halt without poweroff
regardless of /etc/default/halt, which is in contradiction with what the
new version does as you just explained.
--
Pierre Ynard
you think of this? There seems to be a conflict of
opinion here.
--
Pierre Ynard
e boot sequence, and tmpfs /run and
/var/run are long standard now, so I'm downgrading this to wishlist.
Michael, how hard does systemd-tmpfiles depend on systemd, and what
would you think of splitting it into a separate binary package?
--
Pierre Ynard
conclusive and
isn't likely to get anywhere further now, but I wanted to check with
you. Shall we close this?
--
Pierre Ynard
assigning.
I'm sorry for your troubles :(
--
Pierre Ynard
ers 10 years ago, but I wanted to know, what are your reasons
behind this opinion now?
Surely the technical hindrances with the init script aren't a blocker
anymore like they were back then? Are you opposed to adding LSB headers
to cleanly stop udev in runlevel 1 and restart it in runlevel 2? Why?
--
Pierre Ynard
o
this and complained yet. I don't know how popular reiserfs is now, but
there are certainly systems still using it. My point is also that the
assumption that forcefsck was a proprietary feature of e2fsck, was
wrong, and maybe there are yet others.
--
Pierre Ynard
e
time being.
If that check ever makes it back in some form - a safe form, I'm not
sure there is still a point in making it configurable, beyond preventing
dangerous behavior.
--
Pierre Ynard
. Is
there anything you'd still want us to do in the current context, and if
so, what?
--
Pierre Ynard
reassign 595131 e2fsprogs
unarchive 468821
forcemerge 468821 595131
thanks
> fsck died with exit status 1
This was a generic line printed by the logsave utility shipped by
e2fsprogs. It was already replaced by a less scary and confusing
message, for the reasons you mentioned.
--
Pierre Ynard
similar to #813176 affecting systemd-fsck?
--
Pierre Ynard
the forced slow check.
Also, I can see that at least reiserfsck too honors -f as --force with
the same meaning, so I suppose it should be added alongside ext*.
--
Pierre Ynard
s that will not be mounted.
ext3 and others will be mounted fine by this, changing this line won't
solve the problem.
--
Pierre Ynard
reporting issues.
Thanks,
--
Pierre Ynard
e more selective about
filesystems when writing /etc/fstab?
Or as mentioned in #408954, we could also modify initscripts to add
checks and logic that disable fsck on ubifs by default in some way or
another.
--
Pierre Ynard
/boot.d ] && run-parts /etc/boot.d
Without the `exit 0` at the end anymore now that will return a failure
by default.
--
Pierre Ynard
hat I would suggest.
Regards,
--
Pierre Ynard
dev != st2.st_dev) || (st.st_ino == st2.st_ino) );
--
Pierre Ynard
tecture: any, it does not
> make sense. We could provide file with KERNEL= as part
> of package.
If it was Architecture: all; could you still set up that file in
postinst?
--
Pierre Ynard
reassign 460898 util-linux
thanks
The mountpoint executable is now provided from the util-linux source
(see #753779), so I'm reassigning.
Dear maintainers, would you take a look at this patch adding comments in
the mountpoint.c source code?
Thanks,
--
Pierre Ynard
> > Could it be made a conffile?
>
> Definitely. Could you please file a seperate bug about it?
Filed as #923485
--
Pierre Ynard
Package: initscripts
Version: 2.93-8
Severity: wishlist
Dear Maintainer,
/etc/rc.local is currently installed from a static heredoc in
initscripts.postinst. This isn't very clean, and also old systems don't
benefit from new file versions on upgrades, and nothing handles its
removal on uninstall.
tags 734423 patch
thanks
Attaching patch
--
Pierre Ynard
From 3c6577c04166fa90b7aae9b914dc5a69b2a0042d Mon Sep 17 00:00:00 2001
From: Pierre Ynard
Date: Thu, 28 Feb 2019 12:13:51 +0100
Subject: [PATCH] Fix typos in comments in /etc/network/if-up.d/mountnfs
(Closes: #734423)
---
debian/src
tags 686531 patch
thanks
Attaching patch to fix this
--
Pierre Ynard
From 9f5efa15466ca52ca34f458e66ef153f7512059a Mon Sep 17 00:00:00 2001
From: Pierre Ynard
Date: Thu, 28 Feb 2019 19:51:26 +0100
Subject: [PATCH] Skip bootclean.sh on read-only directories (Closes: #686531)
---
debian/src
Patch in attachment
--
Pierre Ynard
From 33ee6752b8752cedd4cdc3bb936485e3aec14e03 Mon Sep 17 00:00:00 2001
From: Pierre Ynard
Date: Thu, 28 Feb 2019 19:16:48 +0100
Subject: [PATCH] Do not use unsafe `: >` shell command to create files
(Closes: #923478)
As per POSIX standard and d
Package: initscripts
Version: 2.93-8
Severity: normal
Tags: patch
As per the POSIX standard:
http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_14
http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_08_01
The `:` is a "special built-in"
> Thank you very much for your research and patches. Can you please
> re-submit your patches in format, generated by `git format-patch'?
>
> What you included is `git show` output, which is not possible to apply
> directly. Thank you again.
Here you go.
--
Pi
c 23.2-1
initscripts suggests no packages.
-- Configuration Files:
/etc/default/rcS changed [not included]
/etc/default/tmpfs changed [not included]
-- no debconf information
commit 289d898ca707be223d4220951cdc3753d21e2df6
Author: Pierre Ynard
Date: Sun Feb 24 17:29:36 2019 +0100
Remove write-o
op-bootlogd-single'
Other reports featuring missing init scripts can be found in #663587 and
#689992, both originally about different issues. Reports in #689992 only
miss /etc/init.d/stop-bootlogd and /etc/init.d/stop-bootlogd-single, not
/etc/init.d/bootlogd
--
Pierre Ynard
/stop-bootlogd-single, so they would be similar to #768555.
--
Pierre Ynard
s thread seems unrelated and would be a case
similar to #768555.
--
Pierre Ynard
ternative packages tzdata, initscripts,
> libc-bin are installed.
I think there was some confusion from the reporter about the kfreebsd
dependency on libc0.1, which he doesn't seem to be using. Hence this is
just another unexplained case of bootlogd not logging to /var/log/boot.
--
Pierre Ynard
What's the reason for using different paths?
Reassigning to systemd, if you guys want to do something with this.
Thanks,
--
Pierre Ynard
d cons and it's really up to the
admin's choice and the use case for the system.
--
Pierre Ynard
umount -l /tmp` could help mitigate this
and restore the system without having to reboot.
--
Pierre Ynard
would suppose that there is no issue
anymore with recent installations and in case there used to be a bug,
support from squeeze was already dropped so that ship has sailed; and
this can probably be closed.
Thanks,
--
Pierre Ynard
ve
with fsck enabled?
Thanks,
--
Pierre Ynard
etc/fstab? The installer? So that the
respective maintainers could take a look at this. This is apparently not
an issue with initscripts.
--
Pierre Ynard
on. For /run/shm ->
/dev/shm, recreating the /dev/shm directory would fail at first but I
suspect it would succeed in the second call.
--
Pierre Ynard
d in
/var/lib/systemd/random-seed instead, does that one get updated?
> Init: systemd (via /run/systemd/system)
The second reporter does use systemd, so I suppose that would be why the
init script updating /var/lib/urandom/random-seed is never run.
--
Pierre Ynard
he initscript
package, however that was switched to the implementation from util-linux
in 2015 (#753779).
I don't know if util-linux's mountpoint does or still does exhibit the
reported problem, but I think it's best to reassign this.
Thanks,
--
Pierre Ynard
ved.
Nowadays logsave/fsck are never launched in the background, and it seems
to work fine for everyone. #682592 was fixed so I would suggest that we
close this too.
--
Pierre Ynard
> I fail to see any typos in my copy of the script. Can you provide more
> details? Is this the same as #587665 fied in -11?
This was most likely a duplicate of #587665, and without other
information there is nothing we can do. I think we should merge and/or
close this.
--
Pierre Ynard
t; >
> > for dir in /lib/init/rw /tmp /run /run/lock /run/shm; do
> > rm -f "$dir/.clean" "$dir/.tmpfs"
> > done
>
> That's much better, thanks.
This proposed snippet could still be relevant (apart from the part about
/lib/init/rw which doesn't exist anymore).
--
Pierre Ynard
in reassigning the bug either.
Can we close this?
--
Pierre Ynard
t included in syslog.
/var/log/boot doesn't log everything that appears on the screen; and
kernel logs may or may not be displayed depending on their severity and
on system settings, which is also a reason to keep them separate.
--
Pierre Ynard
and then restart in runlevel 2. When I look
at udev's initscript, it seems like it must be much better in that
regard now than what was described 10 years ago.
--
Pierre Ynard
t;base" root which is not the
> kernel standard "rootfs". If / is mounted multiple times, this
> unmounts all "stacked" entries as well. In the "aufs" case I described
> above, this ensures the additional partitions are cleanly unmounted.
As you can see, the issue it describes and meant to address is the same
as here. So I believe this bug can be closed.
--
Pierre Ynard
roduce this, is it confirmed that it happened on more
than one system and even that it really depended on systemd?
Also obviously, /lib/init/rw is not in use anymore so we're not likely
to get new developments.
--
Pierre Ynard
s into sed's clipboard every mount
line that isn't mounted on "/", until it finds one that is, and then it
pastes from the clipboard and outputs all those previous lines and the
current one. In effect, that would only remove all the lines after the
last line mounted on "/".
--
Pierre Ynard
nk that it's the responsibility of the
/etc/init.d/networking script to make sure it doesn't hang. Should this
be reassigned to ifupdown so they get a chance to look at it?
--
Pierre Ynard
't too many possible causes for that apart from
genuine mount mishaps and #818442, so it might just be a variation of
it, involving some NFS code path apparently.
--
Pierre Ynard
> Hence, I refine my proposal -- create /var/log/dmesg as 640
> in initscripts, *only* if it does not already exists. Ignore
> kernel.dmesg_restrict.
Makes sense to me.
--
Pierre Ynard
es under /usr was
> the actual classical Unix way.
And there's also the pattern of packages installing their files into
their own private hierarchy under /opt/// ... much
like what happens under \Program Files\ on Microsoft Windows, the
rationale being, it's well-suited to avoid collisions without central
coordination.
Regards,
--
Pierre Ynard
tionality outside of syslog while it may not be available or suited,
and how to manage that and log permissions. I agree that looking
at kernel.dmesg_restrict can be a cool tradeoff, but that's very
specialized.
--
Pierre Ynard
mesg_restrict
seems to make sense but I'm a bit skeptical. I suppose that the way it
keeps permissions set on it by the admin is both a bug and a feature.
--
Pierre Ynard
dmesg output can now be restricted by the kernel.dmesg_restrict sysctl,
so things are not that simple: see also #867747
--
Pierre Ynard
Package: initscripts
Version: 2.93-5
Severity: normal
Tags: patch
I haven't actually confirmed the bug, but just reading the code I
think there's a problem. After 2.93-4 resolved #851427 by reversing
the /dev/shm -> /run/shm symlink to /run/shm -> /dev/shm, on a system
running a static /dev, the
Package: initscripts
Version: 2.88dsf-59
Severity: normal
Tags: patch
It seems to me that the function run_migrate() in /lib/init/mount-functions.sh
has an inverted check:
> # Create symlink if not already present.
> if [ -L "$OLD" ] && [ "$(readlink "$OLD")" !=
their software into
what they choose, and to fix and improve their software to suit their
needs, is good. Please don't say things that feel as if users and use
cases that don't conform to a given ideology are better off rooted out.
--
Pierre Ynard
opy the files they need by hand, out of packaging,
back into /lib or /bin. Then other people will reply: "They're shooting
themselves in the foot, let them rot and die!" And I don't see that as
any good outcome for the community.
--
Pierre Ynard
and -independent executables?
--
Pierre Ynard
c/sysv-rc/rc /usr/libexec/sysvinit
> +debian/src/sysv-rc/rcS /usr/libexec/sysvinit
You've moved them to /usr/libexec/ that doesn't exist on Debian, not
/lib/init/ ?...
--
Pierre Ynard
Package: udev
Version: 239-13
Severity: wishlist
Dear Maintainer,
I looked for old bug reports about this and was surprised not to
find anything filed against udev. My issue is with this snippet in
/etc/init.d/udev:
if ! ps --no-headers --format args ax | egrep -q '^\['; then
log_warning_msg
Package: nullmailer
Version: 1:2.1-5
Severity: wishlist
Dear Maintainer,
As mentioned in the changelog and discussed in #884980, support for sysvinit
was dropped. For reference, quoting relevant entries:
- changelog.Debian
> * Drop support for sysvinit, due to dropping --daemon divergence
Package: flashplugin-nonfree
Version: 1:3.7
Severity: normal
Dear Maintainer,
The Debian packaging of gnupg has changed significantly, in that gnupg
is now a metapackage pulling in the whole gpg suite. However it is my
understanding that the current implementation of flashplugin-nonfree
only
grade path that could unwittingly leave a user in this
situation? I don't think there was any "gpg" package before in a
release? Perhaps it would be better to put gnupg in Recommends too, is
that possible?
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
nce again, I've made several suggestions that would leave
them installed by default like you mentioned. Nobody has denied that it
would be a positive solution for everybody.
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
it would just honor the policy and
restore the freedom to uninstall them to users who don't want or need
them.
Best regards,
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
packages they need.
I hope this tells you more on what you were asking about, and that this
issue can move forward.
Regards,
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
too to achieve that?
I've enjoyed your maintainership, and I dearly hope we can keep enjoying
such builds of Firefox.
Best regards,
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
),
I'm sorry, but in the first place, how many people have to be convinced
to refrain from a policy violation?
> hicolor-icon-theme must stay a hard dependency IMHO.
Okay, I could live with that, but - the point is still there, why must
there be a hard dependency, when GTK2.0 obviously runs fine h
Hello,
It's been many months. Any news, or beginning of an answer about this?
Laurent, can you shed some light in this thread on why the hard
dependency would be needed? Do you mind if we remove it?
Thanks,
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuil
sonally
so I wouldn't quite know what's best to check exactly.
--
Pierre Ynard
"Une âme dans un corps, c'est comme un dessin sur une feuille de papier."
1 - 100 of 187 matches
Mail list logo
